This project will test the hypothesis that neuronal ensembles have nonlinear deterministic properties. There are two aspects of this hypothesis that will be tested: 1) That neuronal ensemble activity can be characterized and controlled through an analysis of unstable periodic orbits, and 2) because of nonlinearity and coupling, neurons within ensembles will exhibit nonlinear synchrony. The theoretical work on the detection of unstable orbits in experimental data will be extended using nonlinear instead of linear fits to data, and also an improved method of detecting orbits of higher period will be developed. These results will be used to characterize neuronal ensemble behavior from both in vitro hippocampal slices and from human epileptic foci. Such orbit information is naturally suited both to characterize the nonstationarity of the dynamics of a system, and also to control the system, despite nonstationarity. This unstable orbit information will be used to control in vitro neuronal ensembles. Nonlinear systems can also demonstrate synchrony that is very different in quality from the more typical synchronization in linear systems. With the recent confirmation of the presence of nonlinear synchrony in neuronal ensemble dynamics, whether such nonlinear synchrony reveals functional connectivity within a neuronal ensemble will be further explored. This will be done by studying dual intracellular patch clamp recordings of pyramidal cell activity as a function of distance (between impalments). In addition, this nonlinear synchrony work will be extended to in vitro and in vivo data with spatial extent in an effort to define spatio-temporal nonlinear synchrony, and such information will be used to help achieve spatio-temporal control in vitro. The results of this research will alter the way that neuronal dynamics can be characterized and controlled, will provide a means to deal with nonstationarity in the nervous system, will broaden the concept of synchrony in the nervous system, and may lay the theoretical foundation for novel approaches for the control of pathological neuronal ensembles in dynamical diseases such as epilepsy.